Thus, if in one year the number of DNA cold-hit convictions (T) goes up 20 percent, while the database increases twofold (100 percent or N = 2NB), then the efficacy for crime solving as determined by our index would go by about 40 percent from what it was when the database was half its present size. Our index for efficacy of DNA data banks grows when DNA plays a stronger role in cold-hit convictions and shrinks when a larger database does little to improve the conviction rate. This indicator tells us that when all the factors of the index remain constant (P, Q, R, S, and T) while the size of the database grows, the efficacy of the DNA data bank decreases.
Size of a DNA Data Bank Versus Solving Crimes
Why would a larger database not necessarily lead to more crimes being solved more efficiently? As previously noted, it is taken for granted by many police authorities that the larger the database, the more crimes will be solved. As an example, consider the following statement published in the annual report of the U.K. NDNAD for 2005–2006: “The additional CJ [criminal justice] arrestee sample profiles to the NDNAD have brought significant benefits, including direct police savings through speedier investigations, quicker apprehension of offenders, earlier elimination of suspects and greater victim reassurance.”20
There are many reasons that a larger database does not necessarily deliver more crime control. First, as we have seen from the previous exercise, the factor currently limiting database efficacy is the number of crime-scene profiles, not the number of offender or subject profiles. Although numbers of subject profiles have increased exponentially in both the United States and the United Kingdom from 2000 to 2010, the number of crime-scene samples added to the database has risen gradually (see figures 17.3 and 17.4). To improve the efficacy of DNA databases in solving crimes, one has to increase the number of crime-scene DNA profiles loaded into the database in the first place.
This is not an easy undertaking. The proportion of crime scenes where DNA has been recovered has increased since 1988 from 4.5 percent to 17 percent in the United Kingdom.21 However, there are limitations on crime-scene profiling that are likely to prevent this number from getting much higher. First, DNA is most readily obtainable from scenes of violent crimes—crimes where biological evidence is likely to be left behind in the form of blood, semen, or skin cells. The majority of crimes that are committed, however, are property crimes. This is true even for the most serious crimes, which are tallied annually by the FBI in its “Crime Index.” Out of the 11.9 million index offenses reported to the FBI in 2002, 1.4 million constituted violent crimes, while 10.5 million were property crimes.22 To look for DNA where it is not obviously present requires painstaking and costly crime-scene investigation, during which forensic technicians scour the scene looking for trace evidence that may or may not carry the DNA of the perpetrator. Moreover, DNA found in these situations may not be of sufficient quality and quantity to permit testing.
FIGURE 17.3. Diminishing returns: CODIS offender profiles (in millions) relative to forensic profiles. Source: CODIS Combined DNA Index System (FBI brochure); CODIS Program Office, FBI, personal communication. Compiled by Tania Simoncelli, Harry G. Levine, and John Gettman, March 2008. Updated by the authors May 2010.
It is possible that DNA technology will continue to change in ways that will allow us to detect even smaller amounts of DNA. But improved DNA detection will not contribute to solving property crimes, where multiple sources of DNA will be found at the same crime scene, and regardless of the technology it will not be possible to sort out which samples belong to the perpetrator. In fact, a concern with overreliance on DNA in criminal investigations is that by attempting to increase the input of crime-scene samples to generate more “hits,” we may end up generating higher numbers of spurious matches. This may in turn decrease the crime-detection and conviction rates and also runs the risk of leading to more wrongful convictions.
FIGURE 17.4. NDNAD offender profiles (in millions) relative to forensic profiles. The number of subject profiles held on NDNAD includes profiles from arrested individuals as well as volunteers and is higher than the number of individuals represented in the database because some of the profiles are replicates. The National DNA Database Annual Report 2007–09 estimates that 13.5 percent of the subject profiles are replicates, such that the number of individuals in the database was approximately 4.9 million as of March 2009. Similarly, the number of crime-scene profiles reported here do not reflect those that have been removed from the database. As of March 2009 there were a total of approximately 350,000 crime-scene profiles retained on the NDNAD (Annual Report, p. 25). Source: Authors. Date Source: National DNA Database Annual Report 2007–09, http://www.npia.police.uk/en/docs/NDNAD07-09-LR.pdf (accessed May 14, 2010).
Even within the category of violent crime the use of DNA is limited. In most cases of rape the rapist is known to the victim. DNA is seldom useful in those cases, since what is most often in dispute is whether the act was forced or consensual (see chapter 8).
In the overall scheme of solving crimes, Wallace notes, “The number of cases that can be solved using DNA will always be limited by the number of crime scenes from which DNA profiles can be collected and the need for corroborating evidence.”23 She adds that even if the DNA match rate (number of DNA matches per crime-scene sample) were 100 percent (for example, if the database were truly universal), the DNA detection rate (or conviction rate) would never be that high.24
There are other reasons to be skeptical about the notion that a larger database is a better one. When DNA databases are expanded to include samples and profiles taken from petty criminals or arrestees who are not charged with a crime, they are including individuals who are less likely to go on to commit future crimes, especially the types of crimes where DNA can be collected. Some argue that adding arrestees to databases will either hinder the conviction rate or not help at all. In 2005 the police liaison officer of the Scottish DNA Database was quoted as saying, “It is arguable that the general retention of profiles from the un-convicted has not been shown to significantly enhance criminal intelligence or detection.”25
Beyond this, there are ways in which DNA expansions can even interfere with or undermine crime-solving efficiency. DNA database expansions create backlogs of unprocessed crime-scene DNA samples, which may remain in a queue for months or even years. As a result, when serious crimes are committed, the crime-scene DNA samples await their turn to be profiled. In 2003 the National Institute of Justice estimated that more than 350,000 rape and homicide cases awaited DNA testing and that crime labs were deluged with analysis requests for convicted-offender samples.26 If we add to the convicted felons individuals arrested but not charged or those detained by federal agencies, the backlog could inhibit the role of DNA in convicting serious felons. As an example, in 2001 police in Wyoming took a DNA sample from an individual convicted of kidnapping. The sample languished in the laboratory without being processed or uploaded to the national database. When it was finally processed, it matched the profile found at a 1997 murder of a University of Colorado student. This case reveals that when biological samples are taken from too many people and scheduled to be processed and uploaded to the forensic DNA database, fewer crimes are actually solved. This may sound counterintuitive. As P. Solomon Banda explained, “The nation’s DNA tracking system is beset with a huge backlog that could take years to clear. And in the meantime, law enforcement officials say, crimes are going unsolved.”27
DNA expansions also are likely to result in more errors in profiling, labeling, and transcription of samples that will add to administrative costs, needless lawsuits, and human hardships for people falsely accused and perhaps even convicted by DNA. Indeed, errors of these types have already occurred, and some have been attributed to the problem of mounting backlogs.28 The error rate will undoubtedly increase as the backlog of DNA profiles rises. Cross-contamination of samples, which has explained past errors and remains a serious quality-control problem in sequencing DNA samples, is more likely to occur when lab
technicians are under increased pressure to speed up processing.
The larger the database, the more opportunity there is for a person to be falsely accused and incriminated for a crime. As the database grows, so too does its potential use as a resource for criminals to frame others for crimes by obtaining and planting DNA evidence from others. Already there have been cases where criminals have planted or tampered with evidence or paid inmates to take DNA tests as a way of confusing investigators or evading prosecution (see the case of Anthony Turner, discussed in chapter 16). Unfortunately, it is not only criminals who might plant DNA in an attempt to frame someone for a crime; there have been a surprising number of police frame-ups reported over the last 10 to 15 years (see chapter 16 for a discussion of Philadelphia police who pleaded guilty to planting illegal drugs on suspects). William C. Thompson notes, “If your profile is in a DNA database you face a higher risk than other citizens of being falsely linked to a crime. You are at higher risk of false incrimination by coincidental DNA matches, by laboratory error and by intentional planting of DNA.”29
The expansion of DNA databases to include DNA profiles of petty criminals and innocent people requires resources, which may come at the expense of traditional criminal investigation techniques. In other words, a transfer of resources takes place from shoe-leather investigations to DNA profiling and database matches. Jenny Rushlow argues that “all of the resources devoted to DNA testing and database management, which are extremely costly and time-intensive, diverts necessary resources away from other aspects of investigations, like tracking down witnesses, victims and suspects.”30 Rockne Harmon, a senior deputy district attorney from Alameda County, California, has complained that the costs associated with building California’s database as a result of the passage of Proposition 69 have taken away resources that are needed to place officers on the streets.31
Finally, thus far we have been discussing the efficacy of DNA data banking in solving crimes. It has also been argued that DNA data banks will prevent or deter crimes from taking place. The presumed conventional wisdom is that people who are in the database will be deterred from committing a crime out of a fear of getting caught. This argument does not apply to the vast majority of innocent people who have not yet and may never commit any serious crimes. Also, there is no evidence that crime rates have declined as a result of the growth of the DNA databases. Furthermore, there is no reason to think that crimes of passion will be deterred by having one’s profile on a DNA data bank. Finally, sophisticated criminals will not be deterred from committing crimes; rather, they will seek to confound the criminal justice system, for example, by introducing foreign DNA into the crime scene.
We return to an earlier discussion of efficacy. By overextending the function of a reliable technology, we can undermine its efficacy. When DNA profiling and data banking are used in connection with violent crimes, there is ample evidence that they can be efficacious and cost effective for generating suspects and evidence that an individual was at the crime scene. However, when the technology is extended to petty crimes or innocent and suspicionless individuals, the evidence leads to the conclusion that there will be a rapid decline in efficacy; the marginal benefits of loading names into a national database decline rapidly. Moreover, there is no evidence that posting people’s DNA profiles on a national database will deter or prevent crimes.
Chapter 18
Toward a Vision of Justice: Principles for Responsible Uses of DNA in Law Enforcement
I don’t know how we, as members of the community, who are concerned with public safety, can look victims and families in the face knowing we can do things in the lab to help bring perpetrators of serious crimes to justice. To me, it’s unconscionable not to use these methods to solve serious crimes and to prevent future crimes.
—Fred Bieber1
Scientific truth-making . . . is always a social enterprise. . . . As such, even scientific claims are subject to distortion, through imperfections in the very human systems that produced them. In attempting to render justice, the law’s objective should be, in part, to restore to view these potential shortcomings, instead of uncritically taking on board a decontextualized image of science that ignores its social and institutional dimensions. Doing justice, after all, demands a complex balancing of multiple considerations.
—Sheila Jasanoff2
Over the last 20 years we have witnessed an extraordinary explosion in the development and use of DNA technology in the criminal justice system. Initially DNA was relied on as an occasionally useful tool in investigating very serious crimes. In those cases DNA, sought from suspects by way of a court-issued warrant supported by probable cause, was compared with DNA left behind at the scene of a heinous crime. But as we have described throughout this book, this fairly circumscribed use of DNA analysis has given way to a massive and ever-expanding system of collecting and permanently retaining DNA for ongoing investigation and use. Most recently a number of law-enforcement policies, techniques, and practices have expanded to allow police to take DNA from innocent people—people who have never been convicted and in some cases never even suspected of a crime.
As described in chapters 2, 3, 4, and 6, the expansion of DNA databases to arrestees, familial searching, the increasing use of DNA dragnets, and surreptitious DNA collection are all ways in which innocent people are increasingly being brought into the criminal justice system by way of their DNA. The mass expansion of DNA collection to innocent people marks a radical shift in the way in which DNA is used in the criminal justice system, one where DNA is starting to look much more like a surveillance tool than a tool for criminal investigation. Throughout this book we have explored multiple dimensions of these developments, including their impact on crime solving, their role in exonerating the wrongly convicted, their implications for privacy, and their potential for error. But perhaps the overriding issue at stake is whether DNA data banks, now a mainstay in all modern industrial societies, are bringing greater justice to the policing functions of civil society, while minimizing injustice. Will ever more DNA expansion necessarily advance the cause of justice?
The answer may very well depend on how one perceives “justice” and its role in the criminal justice system. When we speak of the “criminal justice system,” it usually refers to a set of practices and institutions of government directed at maintaining social order, defining, deterring, and mitigating crime, and establishing sanctions, penalties, and rehabilitation for those convicted of crimes. The system includes the perpetrators and victims of crime, the falsely accused and wrongfully incarcerated, the courts, the police, prosecutors, the defense bar, the written body of legal precedents established in case law, and the complex tapestry of regulations and rules that set standards of behavior and accountability within the criminal justice subsystems. But how does “justice” enter into “criminal justice”?
Concepts of Justice in Criminal Law
In law enforcement and its ancillary institutions the concept of justice is expressed in several distinct forms. Retributive justice (also punitive justice) is based on the idea that the guilty must be apprehended and punished for their crimes. Without some form of retribution, it is often argued, there would be no deterrence against violating the laws of civil society and committing injustices against persons or the state. The victims of crimes are often the most vocal about seeking retribution. Retributive justice draws from the ethics of “an eye for an eye” or the philosophy of the “just desert,” namely, that “you reap what you sow.”
The concept of restitutive (restorative) justice signifies that the victim’s losses shall be restored or that some effort is made to pay restitution for one’s crimes. In the criminal justice system restitutive justice is introduced in particular cases where a convicted felon is obliged to repay the victim with some form of restitution for causing diminished capacity or to repay the state for crimes against property. Tort law is premised on elements of restitutive and retributive justice, for example, when courts award pu
nitive and compensatory damages for losses. Some states provide financial relief for those wrongfully convicted and incarcerated for a crime, where the amount of compensation depends on the length of the term served.
Procedural justice seeks to ensure that those charged with a crime obtain adequate counsel and are afforded equal opportunity (compared with anyone else charged with a similar crime) to prove their innocence. The term “due process” or “due process of law” implies that the government must respect all the legal rights that are owed to an individual, and that a law shall not be unreasonable, arbitrary, or capricious. The constitutional guarantee of due process of law is found in the Fifth and Fourteenth Amendments to the U.S. Constitution. The Fifth Amendment prohibits the federal government from arbitrarily or unfairly depriving individuals of their basic rights to life, liberty, and property; the Fourteenth Amendment applies these same limitations to the states. The Fourteenth Amendment has been interpreted by the U.S. Supreme Court to incorporate the basic civil liberties protections of the Bill of Rights, so that those protections also apply to the states, as well as to the federal government. Injustices include unwarranted intrusion into one’s privacy, incarceration without being charged of a crime, and the failure to afford an individual charged with a felony the right to a trial by an impartial jury of one’s peers.
Procedural justice also implies that justice for an individual is associated with fair and equal treatment under the law. This includes the idea of “fair sentencing.” It would be unjust if some people received disproportionately harsh sentences for the same crime. Procedural justice also includes the ideas of “presumed innocence” and “burden of proof.” Innocence is the default state unless one can demonstrate otherwise. In addition, it is the state’s burden to demonstrate guilt rather than the suspect’s burden to prove innocence.
Genetic Justice Page 39